Identifying Underlying Dimensions in Spouses' Evaluations of Fairness in the Division of Household Labor

Herbert L. Smith, Constance Gager, S. Philip Morgan

Research output: Contribution to journalArticleResearchpeer-review

36 Citations (Scopus)

Abstract

Using the first wave of the National Survey of Families and Households (NSFH-I), we analyze the joint distribution of wives' and husbands' answers regarding whether the division of household chores is "fair." Responses comprise a 5-point scale, from "very unfair to me" to "very unfair to her/him" with "fair to both" as a middle category. It is well known that wives are more likely than husbands to report that the division of household labor is "unfair to me," but this does not begin to exploit the information contained in these paired (wife/husband) data. Correlational measures suggest that agreement between wives and husbands is low, but the absolute levels of agreement are high, since, in the majority of couples, both spouses say the situation is "fair to both." The association is well captured by a restricted version of the log-linear model for quasi-symmetry. This model suggests a two-dimensional latent trait model for the pattern of couple-specific responses. Additional analyses indicate that the NSFH-I "fairness" item is a convolution of responses along two dimensions - one reflecting "relative fairness" (e.g., fair to one spouse compared to the other's contribution), the other a tendency on the part of many couples to report "fair to both" regardless of the "true" level of fairness between the wife and husband. This multidimensionality shows that there are difficulties in analyzing this item that have been hitherto unremarked upon in the literature. We argue that the latent trait perspective suggests that evaluations of equity need to be interpreted in terms of characteristics of couples, or marriages, not just in terms of the characteristics of wives and/or husbands.

Original languageEnglish
Pages (from-to)305-327
Number of pages23
JournalSocial Science Research
Volume27
Issue number3
DOIs
StatePublished - 1 Sep 1998

Fingerprint

fairness
husband
spouse
wife
labor
evaluation
linear model
equity
marriage

Cite this

@article{6c65a94c1a864ee78e0242a75aac269e,
title = "Identifying Underlying Dimensions in Spouses' Evaluations of Fairness in the Division of Household Labor",
abstract = "Using the first wave of the National Survey of Families and Households (NSFH-I), we analyze the joint distribution of wives' and husbands' answers regarding whether the division of household chores is {"}fair.{"} Responses comprise a 5-point scale, from {"}very unfair to me{"} to {"}very unfair to her/him{"} with {"}fair to both{"} as a middle category. It is well known that wives are more likely than husbands to report that the division of household labor is {"}unfair to me,{"} but this does not begin to exploit the information contained in these paired (wife/husband) data. Correlational measures suggest that agreement between wives and husbands is low, but the absolute levels of agreement are high, since, in the majority of couples, both spouses say the situation is {"}fair to both.{"} The association is well captured by a restricted version of the log-linear model for quasi-symmetry. This model suggests a two-dimensional latent trait model for the pattern of couple-specific responses. Additional analyses indicate that the NSFH-I {"}fairness{"} item is a convolution of responses along two dimensions - one reflecting {"}relative fairness{"} (e.g., fair to one spouse compared to the other's contribution), the other a tendency on the part of many couples to report {"}fair to both{"} regardless of the {"}true{"} level of fairness between the wife and husband. This multidimensionality shows that there are difficulties in analyzing this item that have been hitherto unremarked upon in the literature. We argue that the latent trait perspective suggests that evaluations of equity need to be interpreted in terms of characteristics of couples, or marriages, not just in terms of the characteristics of wives and/or husbands.",
author = "Smith, {Herbert L.} and Constance Gager and Morgan, {S. Philip}",
year = "1998",
month = "9",
day = "1",
doi = "10.1006/ssre.1998.0624",
language = "English",
volume = "27",
pages = "305--327",
journal = "Social Science Research",
issn = "0049-089X",
publisher = "Academic Press Inc.",
number = "3",

}

Identifying Underlying Dimensions in Spouses' Evaluations of Fairness in the Division of Household Labor. / Smith, Herbert L.; Gager, Constance; Morgan, S. Philip.

In: Social Science Research, Vol. 27, No. 3, 01.09.1998, p. 305-327.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - Identifying Underlying Dimensions in Spouses' Evaluations of Fairness in the Division of Household Labor

AU - Smith, Herbert L.

AU - Gager, Constance

AU - Morgan, S. Philip

PY - 1998/9/1

Y1 - 1998/9/1

N2 - Using the first wave of the National Survey of Families and Households (NSFH-I), we analyze the joint distribution of wives' and husbands' answers regarding whether the division of household chores is "fair." Responses comprise a 5-point scale, from "very unfair to me" to "very unfair to her/him" with "fair to both" as a middle category. It is well known that wives are more likely than husbands to report that the division of household labor is "unfair to me," but this does not begin to exploit the information contained in these paired (wife/husband) data. Correlational measures suggest that agreement between wives and husbands is low, but the absolute levels of agreement are high, since, in the majority of couples, both spouses say the situation is "fair to both." The association is well captured by a restricted version of the log-linear model for quasi-symmetry. This model suggests a two-dimensional latent trait model for the pattern of couple-specific responses. Additional analyses indicate that the NSFH-I "fairness" item is a convolution of responses along two dimensions - one reflecting "relative fairness" (e.g., fair to one spouse compared to the other's contribution), the other a tendency on the part of many couples to report "fair to both" regardless of the "true" level of fairness between the wife and husband. This multidimensionality shows that there are difficulties in analyzing this item that have been hitherto unremarked upon in the literature. We argue that the latent trait perspective suggests that evaluations of equity need to be interpreted in terms of characteristics of couples, or marriages, not just in terms of the characteristics of wives and/or husbands.

AB - Using the first wave of the National Survey of Families and Households (NSFH-I), we analyze the joint distribution of wives' and husbands' answers regarding whether the division of household chores is "fair." Responses comprise a 5-point scale, from "very unfair to me" to "very unfair to her/him" with "fair to both" as a middle category. It is well known that wives are more likely than husbands to report that the division of household labor is "unfair to me," but this does not begin to exploit the information contained in these paired (wife/husband) data. Correlational measures suggest that agreement between wives and husbands is low, but the absolute levels of agreement are high, since, in the majority of couples, both spouses say the situation is "fair to both." The association is well captured by a restricted version of the log-linear model for quasi-symmetry. This model suggests a two-dimensional latent trait model for the pattern of couple-specific responses. Additional analyses indicate that the NSFH-I "fairness" item is a convolution of responses along two dimensions - one reflecting "relative fairness" (e.g., fair to one spouse compared to the other's contribution), the other a tendency on the part of many couples to report "fair to both" regardless of the "true" level of fairness between the wife and husband. This multidimensionality shows that there are difficulties in analyzing this item that have been hitherto unremarked upon in the literature. We argue that the latent trait perspective suggests that evaluations of equity need to be interpreted in terms of characteristics of couples, or marriages, not just in terms of the characteristics of wives and/or husbands.

UR - http://www.scopus.com/inward/record.url?scp=0000134322&partnerID=8YFLogxK

U2 - 10.1006/ssre.1998.0624

DO - 10.1006/ssre.1998.0624

M3 - Article

VL - 27

SP - 305

EP - 327

JO - Social Science Research

JF - Social Science Research

SN - 0049-089X

IS - 3

ER -